How will coding agents change your code? While at Build, Carl and Richard chatted with Scott Hunter about the announcements around coding agents at the keynote. Scott talks about the agent mode available in Visual Studio Code - and now in Visual Studio! Agent mode allows the LLM to evaluate the code across an entire solution, not just the file you're currently looking at. You can create a workflow where GitHub issues are assigned to the agent, which then generates code and provides a pull request for evaluation. The agents are here and helping us do more!

What if you could use ElasticSearch serverless? While at Build, Carl and Richard chatted with Ken Exner about the new announcements around Elastic providing serverless storage and search! Ken talks about paying for only the data you move and store with serverless, rather than needing to operate any infrastructure for Elastic. The conversation digs into the potential of Elastic in Azure AI Foundry to provide ultra-fast access to current company data for your LLM implementations. Elastic did vector databases before LLMs made them essential for RAG - and you can take advantage of it!

Do you understand how networking works in C#? Carl and Richard chat with Chris Woody Woodruff about his new book on networking with C#. Chris runs down the fundamentals of networking and then discusses the different approaches readily available in the C# world, including web sockets, gRPC, SignalR, and many more! The conversation also turns to the upcoming QUIC standards built into HTTP/3 that should simplify networking. Sure, you could go with the defaults, but why not explore all the options!

How do you secure browser-based frontends with ASP.NET Core backends? Carl and Richard discuss the Backend for Frontend (BFF) Security Framework with Erwin van der Valk. Erwin talks about Sam Newman's BFF Pattern and how it helps deal with the diversity of clients, including web, desktop, and mobile, to work with a common backend. OAuth 2.0 is capable of dealing with this complexity, but there are many moving parts, and that's where the security framework can help!

How do you balance architecture and code? Carl and Richard talk to Steve Smith about various architectural strategies and the swing back-and-forth against over-designing architecture and getting code written. Steve talks about how architecture changes depending on the size and number of teams, how the latest tools can help with architectural choices, and the challenge of effective refactoring when things need to change. Lots of great conversation!

Open Source Maintainers are burning out or going commercial - how do we solve this? Carl and Richard chat with Rob Mensching about his work to create the open source maintenance fee through GitHub. Rob talks about the common problem of single maintainers getting buried under issues and demands of consumers for a project. Recognizing that most people cannot contribute to the project, a maintenance fee helps support the maintainer in a low friction way for everyone involved. Check out the links to get started!

How can a large language model help your organization answer government RFPs? Carl and Richard talk to Vishwas Lele about his startup pWin, as in proposal win. Vishwas talks about being a year into the startup and his deeper understanding of how AI technologies can augment skilled operators to produce better quality products in less time, including responding to RFPs. The conversation digs into tuning the LLM to focus on the data relevant to each section of the RFP so that the operator can interact with the tool and build better responses!

Can you integrate performant audio-video into your .NET application? Carl and Richard talk to Elias Puurunen about his work at Tractus Events, where he uses the NDI protocols to bring real-time audio and video streams into his C# application. Elias talks about the power of P/Invoke to access the underlying libraries for controlling video streams, including utilizing NVidia GPUs for extremely fast encoding and decoding. You could write this code in C++, but why?

Ready to build an agentic AI in .NET? Carl and Richard talk to Spencer Schneidenbach about his work using large language models to enhance customer interactions in healthcare. Spencer discusses using the LLMs to summarize customer conversations to identify topic areas, sentiment, and other concerns. He digs into how Microsoft's Semantic Kernel makes connecting an OpenAI model to your APIs easy, fetching information and creating a context for testing reliability and consistency with these models. Check out the links for some great tools to help make your AI apps with .NET!

How do you measure the quality of a large language model? Carl and Richard talk to Dr. Jodie Burchell about her work measuring large language models for accuracy, reliability, and consistency. Jodie talks about the variety of benchmarks that exist for LLMs and the problems they have. A broader conversation about quality digs into the idea that LLMs should be targeted to the particular topic area they are being used for - often, smaller is better! Building a good test suite for your LLM is challenging but can increase your confidence that the tool will work as expected.